Elsevier

Psychoneuroendocrinology

Volume 24, Issue 1, 1 January 1999, Pages 69-84
Psychoneuroendocrinology

1998 Curt P. Richter Award
Pregnancy, the postpartum, and steroid hormones: effects on cognition and mood

https://doi.org/10.1016/S0306-4530(98)00044-4Get rights and content

Abstract

The effects of pregnancy on cognition and mood were examined using a repeated-measures design. Nineteen women, average age 33, were tested with a comprehensive neuropsychological battery during their last 2 months of pregnancy and again within 2 months of delivery. Blood samples were obtained from all subjects and assayed for a variety of steroid hormones implicated in cognitive and mood functioning. Most participants also completed several self-report measures of mood. In comparison with performance after delivery, women showed significantly more impairment in aspects of verbal memory during pregnancy and also tended to report more negative mood states. Memory deficits were not explained by mood disturbances. No hormone assayed consistently related to cognitive performance during pregnancy. During pregnancy, higher levels of progesterone (P) were associated with greater mood disturbances and higher levels of dehydroepiandrosterone (DHEA) with better mood. After delivery, testosterone (T) was strongly and consistently associated with greater reported mood disturbances. Our results confirm a peripartal memory deficit, which cannot be explained by the dramatic rise in circulating steroid hormones, or by mood status during pregnancy. Steroidal hormones, namely P, DHEA and T, appear to play a role in mood disturbances during, and after, pregnancy. Studies beginning earlier in pregnancy and continuing for an extended period of time after delivery are needed to confirm and expand these observations.

Introduction

Cognitive and mood changes during pregnancy and in the postpartum period are frequently self-reported and have been documented empirically. Mood, as it relates to pregnancy, has been addressed primarily in relation to postnatal depression. However, available estimates of negative affective symptomatology during pregnancy suggest a sizable proportion of women are affected (Nott et al., 1976, O’Hara et al., 1984, O’Hara, 1986). Mood changes in the postnatal period include mild forms of dysphoria, typically referred to as the blues (Harris, 1980, Pritchard and Harris, 1996), and anxiety and tearfulness (Kennerly and Gath, 1986). Mild to moderate depression is ≈3× higher in the month following delivery than in non-gestational controls (O’Hara et al., 1990).

While negative affective state has been investigated in relation to a number of sociodemographic variables, postnatal depressed mood has been related to the hormonal changes of the perinatal period. Nott et al. (1976) report a correlational trend for greater antenatal irritability with higher estrogen levels, as well as a greater likelihood for subjects with larger progesterone (P) declines to rate themselves as depressed postnatally. O’Hara (1986) also examined P and mood from the second trimester of pregnancy to the 9th postpartum week but failed to find a similar correlation. However, estrogen withdrawal was identified as a possible cause of postpartum blues. A prospective study of primiparous women, which assessed mood and concentrations of P and cortisol antenatally and in the month following parturition, found that women with maternity blues had significantly higher antenatal P concentrations and lower postnatal concentrations than women without blues (Harris et al., 1994). In contrast, Heidrich et al. (1994) reports no significant difference in estradiol (E2) or P levels between women with and without postpartum blues.

Estrogen levels may be associated with changes in mood in non-pregnant women. The use of ERT is associated with less depression in the post-menopause (Sherwin and Gelfand, 1985, Ditkoff et al., 1991). Hormonal fluctuations across the menstrual cycle have been inconsistently associated with changes in mood (Halbreich et al., 1986, Phillips and Sherwin, 1992b, Redei and Freeman, 1995, Fink et al., 1996).

Fewer studies have analyzed cognitive changes during and after pregnancy. Sharp et al. (1993) found deficits in the recall of word lists in both primigravid and multigravid pregnant women when compared with non-pregnant women. This deficit was greater for incidental recall than when the women were explicitly instructed to remember items. A similar pattern was found for pregnant women in all trimesters of pregnancy. Eidelman et al. (1993) found that women in the 3rd month of a high risk pregnancy performed worse than nonpregnant controls on the ability to recall a short passage of prose, but not on a visual memory task. In a separate group of women with normal pregnancies, deficits in both passage recall and visual memory were reported during the 1st postpartum day but not during days two and three. Silber et al. (1990), testing women in the 36th week of pregnancy, within 3 days after delivery, and at 3, 6 and 12 months postpartum, found no differences between pregnant women and controls during pregnancy, at the immediate postpartum or at the 3 month testing. However, the women who had been pregnant showed significant increases in performance on word list learning and reaction time at the 6 and 12 month testings, when compared with the controls. The authors interpret this as indicative of a peripartal cognitive impairment. In a longitudinal study, Keenan et al. (1997), find explicit memory deficits, using paragraph recall, only in the third trimester, when pregnant women are compared with controls. Subjective reports of cognitive dysfunction during pregnancy have been noted. Poser et al. (1986) found symptoms such as forgetfulness, disorientation, confusion, and reading difficulties to be common among pregnant professional women. Subjective memory complaints during pregnancy have also been associated with impairments in implicit memory (Brindle et al., 1991). None of the studies on cognitive functioning have simultaneously evaluated mood, which is essential to determine if cognitive changes are reflective of changes in mood.

The mechanism for any cognitive changes associated with pregnancy remains to be determined. Silber et al. (1990) found no consistent associations between cognitive changes and changes in oxytocin levels. In studies not involving pregnancy, there is compelling support for an effect of steroid hormones on both cognitive and mood functioning. Among women, fluctuations in circulating estrogens, from either endogenous or exogenous sources, are associated with specific cognitive profiles. This has been most widely reported for administration of estrogen replacement to women after either surgical (Sherwin, 1988, Phillips and Sherwin, 1992a) or natural menopause (Kampen and Sherwin, 1994, Robinson et al., 1994). Studies also suggest that the estrogen deficiency that results with the use of gonadotropin-releasing-hormone (GnRH) agonists is associated with decreased memory (Kortepeter et al., 1992, Newton et al., 1996), an effect that is reversed by add-back estrogen therapy (Sherwin and Tulandi, 1996). Endogenous fluctuations in estrogen associated with the menstrual cycle have also been associated with cognitive changes such that specific cognitive skills are elevated at times during the cycle when estrogen levels are high (Hampson and Kimura, 1988, Hampson, 1990).

Other hormones have been associated with cognitive and mood status including P (Freeman et al., 1993, Baker et al., 1995), testosterone (T) (Janowsky et al., 1994, Sands and Studd, 1995), dehydroepiandrosterone (DHEA) (Morales et al., 1994, Wolkowitz et al., 1997), and cortisol (Carroll et al., 1981, Lupien et al., 1994). These hormones also change dramatically during pregnancy and with parturition.

In an effort to better understand both the cognitive and mood effects of pregnancy and the postpartum, we evaluated a group of women during the last month of pregnancy and again within 2–6 weeks after delivery. These times were chosen to allow for a determination of the effects of late pregnancy, when hormones are elevated, in comparison with performance in the postpartum, after the initial trauma of labor has subsided, while steroid hormones are still greatly reduced. We utilized a comprehensive battery of neuropsychological tests that have widely used norms available. Tests of mood are included to allow for evaluation of mood during and after pregnancy and to be able to determine if any cognitive changes are independent of mood. A range of steroid hormones, known to have effects on cognition and mood, were evaluated in an attempt to elucidate the mechanism underlying any effects of pregnancy and the postpartum on cognition or mood.

Section snippets

Subjects

Participants were recruited through private practice offices of Los Angeles area gynecologists and through birthing classes conducted at a major Los Angeles hospital. All subjects were primary English-speakers and had uncomplicated pregnancies. Twenty-five subjects were enrolled into the study and 19 completed the neuropsychological tests on both occasions and are included in current analyses. The average age of the subjects was 33.1 years (±4.7) and average education was 16.0 years (±1.3).

Pregnancy and cognition

Descriptive and statistical information for all cognitive tests is provided in Table 2. Subjects showed no change in general intelligence level during and after pregnancy, performing slightly above the mean at both testing occasions, on the TONI. On the CVLT, subjects performed significantly worse when pregnant on the number of items recalled after the fifth trial (F=13.23, p=.002). Subjects also showed a significantly lower learning slope across the five learning trials when pregnant (F=5.13, p

Discussion

These findings confirm previous reports of peripartal cognitive deficits (Poser et al., 1986, Silber et al., 1990, Brindle et al., 1991, Eidelman et al., 1993, Sharp et al., 1993, Keenan et al., 1997). Specifically, cognitive performance was worse, in certain domains, during pregnancy than relatively shortly after delivery, although performance continued below expected normative values on aspects of verbal memory. Subjects did not report greater mood disturbances after delivery, rather there

Acknowledgements

The impetus for this study came from discussions with Caleb E. Finch, Ph.D., and his seminal work on reproduction and senescence. Portions of this study were presented at the International Neuropsychological Society Meeting, Orlando FL, February, 1997 and at the Society for Gynecologic Investigations, San Diego CA, March, 1997.

References (57)

  • J.Z. Scott et al.

    A double-antibody radioimmunoassay for serum progesterone using progesterone-3-oximino-[125I]-histamine as radioligand

    Steroids

    (1978)
  • B.B. Sherwin

    Estrogen and/or androgen replacement therapy and cognitive functioning in surgically menopausal women

    Psychoneuroendocrinology

    (1988)
  • B.B. Sherwin et al.

    Sex steroids and affect in the surgical menopause: a double-blind, cross-over study

    Psychneuroendocrinology

    (1985)
  • M. Silber et al.

    Temporary peripartal impairment in memory and attention and its possible relation to oxytocin concentration

    Life Sciences

    (1990)
  • N. Sitaram et al.

    Choline: selective enhancement of serial learning and encoding of low imagery words in man

    Life Sciences

    (1978)
  • O.M. Wolkowitz et al.

    Dehydroepiandrosterone (DHEA) treatment of depression

    Biological Psychiatry

    (1997)
  • E.R. Baker et al.

    Efficacy of progesterone vaginal suppositories in alleviation of nervous symptoms in patients with premenstrual-syndrome

    Journal of Assisted Reproduction Genetics

    (1995)
  • Benton, A., deS Hamsher, K., Varney, N.R., Spreen, O., 1983. Judgment of Line Orientation. Oxford University Press, New...
  • P.M. Brindle et al.

    Objective and subjective memory impairment in pregnancy

    Psychology and Medicine

    (1991)
  • Brown, L., Sherbenou, R.J., Johnsen, S.K., 1990. Test of Nonverbal Intelligence Pro-Ed, Austin,...
  • B.J. Carroll et al.

    A specific laboratory test for the diagnosis of melancholia

    Archives of General Psychiatry

    (1981)
  • Delis, D.C., Kramer, J.H., Kaplan, E., Ober, B.A., 1987. The California Verbal Learning Test. The Psychological...
  • Derogatis, L.R., 1983. SCL-90-R Manual II. Clinical Psychometric Research, Towson,...
  • E.C. Ditkoff et al.

    Estrogen improves psychological function in asymptomatic postmenopausal women

    Obstetrics and Gynecology

    (1991)
  • J.F. Dorgan et al.

    Relations of serum DHEA, DHEAS, and 5-androstene-3β,17β-diol to risk of breast cancer in postmenopausal women

    Cancer Epidemiology and Biomarkers Prevention

    (1994)
  • D.A. Drachman

    Memory and cognitive function in man: does the cholinergic system have a specific role?

    Neurology

    (1977)
  • A.I. Eidelman et al.

    Cognitive deficits in women after childbirth

    Obstetrics and Gyncology

    (1993)
  • G. Fink et al.

    Estrogen control of central neurotransmission-effect on mood, mental state, and memory

    Cellular Molecular Neurobiology

    (1996)
  • Cited by (184)

    • Fronto-parietal alpha ERD and visuo-spatial attention in pregnant women

      2023, Brain Research
      Citation Excerpt :

      Evaluating whether accuracy in visuo-spatial attention differs between pregnant and non-pregnant women therefore seems to require either a bigger sample size or a more challenging attention task. The inconsistencies between self-reports and standardised attention tests are already well-documented in the literature (Crawley et al., 2008; Crawley et al., 2003; Logan et al., 2014, Farrar et al., 2014; Henry and Sherwin, 2012, Galen Buckwalter et al., 1999; Raz, 2014). For example, Crawley et al. (2003) reported that pregnant women complained about mild impairments in their focused and divided attention ability compared to non-pregnant controls, while these self-reported deficits were not revealed in standardised objective tests, like the “Stroop task” and the “Trail making task”.

    View all citing articles on Scopus
    View full text